In the field of light measurement and densitometry, the need to convert from an electrical voltage proportional to light intensity to another signal proportional to optical density has long been recognized. One approach in the prior art has been to obtain a sensor voltage signal representing intensity of transmitted or reflected light and convert this analog signal to digital form. The digital value is then used to enter a stored LUT of intensity and density values. The digital density value corresponding to the digital intensity value is read from the LUT.
U.S. Pat. No. 5,117,119 discloses an automatic gain selection electronic circuit, along with a second LUT to obtain high accuracy and resolution over an increased range of large densities. The first (or "base") LUT contains density values corresponding to an analog-to-digital converter output for the lowest gain. The second (or "range") LUT is much smaller than the first LUT and contains the relative density corresponding to each available gain. It provides the additional density output associated with the gain selected. The two LUT outputs are summed to obtain the actual density measurement.
The three ranges illustrated in U.S. Pat. No. 5,117,119 are divided by two threshold values in a 10:1 ratio. Thus two ranges have 10:1 max-to-min light or voltage input ratios, and the third range (used for lowest light intensities or highest density) may have arbitrarily small input light or voltage level. The illustration in U.S. Pat. No. 5,117,119 shows accuracy of 0.01 density units, but requires a 10-bit analog-to-digital converter to do so. In turn, the 10-bit analog-to-digital converter requires a large "base" LUT of 2.sup.10 =1024 entries. A major limiting factor in accuracy is the analog-to-digital converter resolution. Worst-case density resolution for each range comes at the high-density (low light intensity) end of the range, where the analog-to-digital converter resolution, i.e., one count, corresponds to the largest density increment. U.S. Pat. No. 5,117,119 uses multiple analog threshold voltage levels for comparison to the light sensor voltage signal. Low levels of electrical noise and circuit variability could degrade the comparator accuracy and reliability for the low-voltage thresholds.